Hyperfine fields in UFe5Sn compound

PROCEEDINGS
NUKLEONIKA 2007;52(Supplement 1):S67−S69
Hyperfine fields in UFe5Sn compound
Dariusz Satuła,
Krzysztof Szymański,
Vinh H. Tran,
Ludwik Dobrzyński
Abstract. Local hyperfine interactions on Fe nuclei in UFe5Sn compound were determined in Mössbauer experiment.
The analysis of the Mössbauer spectra measured at room temperature reveals the presence of two doublets only.
Absorption line width indicates that neither distribution of quadrupole splitting nor isomer shift exists in the alloy. The
measurements at 13 K show the presence of four magnetic components. Because the magnetic dipole and electric
quadrupole interactions are present, full Hamiltonian was used to determine the hyperfine interactions for each iron
site. Electric field gradient determined at low temperatures agrees with quadrupole splitting observed at room
temperature. Absorption areas of the components at low and at high temperatures correspond well to the occupation of
the crystal sites by Fe atoms.
Key words: Mössbauer studies • uranium intermetallic • magnetic intermetallic
Introduction
D. Satuła, K. Szymański
Institute of Experimental Physics,
University of Białystok,
41 Lipowa Str., 15-424 Białystok, Poland,
Tel.: +48 85 7457242, Fax: +48 85 7457223,
E-mail: [email protected]
V. H. Tran
W. Trzebiatowski Institute of Low Temperature
and Structure Research, Polish Academy of Sciences,
50-950 Wrocław, P.O. Box 937, Poland
L. Dobrzyński
Institute of Experimental Physics,
University of Białystok,
41 Lipowa Str., 15-424 Białystok, Poland
and The Andrzej Sołtan Institute for Nuclear Studies,
05-400 Otwock-Świerk, Poland
Received: 20 June 2006
Accepted: 22 September 2006
New intermetallic compound UFe5Sn, discovered by
A. P. Gonçalves et al. [3], crystallizes in orthorhombic
CeCu5Au-type structure (space group Pnma). The unit
cell contains 4 U atoms located in 4c site, 4 Sn atoms in
4c, 12 Fe atoms in 4c (Fe2, Fe3, Fe4) and 8 Fe atoms
in 8d (Fe1) site [3]. The magnetization measurements
carried out on a powder sample show a weak ferromagnetic behavior at room temperature and two magnetic
anomalies at 248 K and 180 K [3]. The latter transition
is magnetic field dependent and disappears at B > 1 T.
Below 248 K magnetization strongly increases with
decreasing temperature. More precise experiments on
a single crystal [4] show that the transition at 248 K is
associated with ferromagnetic ordering of iron magnetic
moments along c-axis, while anomalies observed at
180 K with reorientation Fe moments towards to b-axis.
This conclusion is deduced partially from Mössbauer
measurements. However, the description of the spectra
presented in [4] in our opinion does not fit the experimental data so well. We think that the main reason of
not good agreement between the experimental and
theoretical Mössbauer spectra is the use of the first order
perturbation in the presence of mixed hyperfine interactions. Since the electric quadrupole interactions are
not small with respect to the magnetic dipolar ones,
a full Hamiltonian should be used for correct description of the Mössbauer spectra. The aim of our study is
D. Satuła et al.
S68
determination of the hyperfine interactions of iron
atoms located at different sites in the UFe5Sn compound.
Sample preparation and data handling
The UFe5Sn crystal was grown by the Czochralski
method. Our sample crystallizes in the same structure
as reported [3] and the lattice parameters, within the
experimental errors, do not differ greatly from [3]. The
powdered sample studied in the work was crushed from
a single crystal, and was mixed with the LiCO3 and epoxy
glue. Such preparation results in the random distribution of the grains, so-called texture free sample. The
sample thickness was 12 mg UFe5Sn per cm2.
The Mössbauer measurements at T = 13 K were
carried out in a close cycle refrigerator equipped with an
antivibration shroud. The source of radiation was 57Co
in Cr matrix. The velocity scale was calibrated with
respect to α-Fe at room temperature.
All spectra were analysed by commercial Mössbauer
package “Normos” using transmission integral [5].
Recoilless fraction of Fe atoms in UFe5Sn was assumed
to be the same as in α-Fe [2]. The resulting values of
effective thickness of the absorber were estimated to be
τ = 2.07 ± 0.11 at room temperature and τ = 2.41 ± 0.11
at T = 13 K.
The spectra measured at T = 13 K were analyzed
using full Hamiltonian approach for the texture free
sample. Axial symmetry of the electric field gradient
was assumed (asymmetry parameter η = 0) for all local
environments of Fe atoms.
Table 1. The hyperfine parameters obtained from the
spectrum measured at room temperature (eQ − electric
quadrupole moment, Vzz − z component of electric field
gradient, IS − isomer shift, I − intensities of the components
and Γa − full width of the Lorentzian lines at half maximum.
The width of the source was determined from calibration
experiment Γs = 0.17 mm/s)
|eQVzz/2|
(mm/s)
IS
(mm/s)
I
(%)
Γa
(mm/s)
D(1)
0.54 ± 0.01
−0.06 ± 0.01
84 ± 4
0.10 ± 0.01
D(2)
0.44 ± 0.02
0.11 ± 0.01
16 ± 4
0.10 ± 0.01
The Mössbauer spectrum measured at room temperature was fitted by two doublets with the same values of
width of the Lorentzian lines (D(1), D(2)). The result
of the best fit is shown in Fig. 1 by the thick line, while
parameters of the two components are listed in
Table 1. The ratio of the areas of the two doublets is
close to four, consistent with the assumption that doublet
D(1) is created by 8 Fe atoms at 8d and 8 Fe atoms at 4c
sites, while D(2) doublet is created by 4 Fe atoms at
4c site. The difference in the isomer shift (IS) between
D(2) and D(1) doublets is 0.18 mm/s. Because the Sn
atoms located in the nearest neighborhood (NN) of Fe
increase its IS [1], one may conclude that D(2) doublet
results from the Fe2 site only, which has three Sn atoms,
and being their NN located at distances ranging
between 2.61−2.93 Å [3].
Despite four different local Fe sites in the unit cell,
the room temperature spectra can be precisely fitted
by two doublets only (see Fig. 1). The absorption lines
are not broadened (their widths are equal to the widths
of the lines in calibration spectrum of α-Fe). We thus
conclude that hyperfine interactions for the Fe1, Fe3
and Fe4 sites, which differ in local symmetry, result in
the single doublet D(1).
The Mössbauer spectrum measured at T = 13 K is
shown in Fig. 2. The shape of spectrum can be fitted by
a sum of four magnetic components S(n) (n = 1,2,3,4)
and a single doublet (D), all with very sharp lines (Γa =
0.13 ± 0.01 mm/s). Thick lines in Fig. 2 represent the
best fit. The parameters obtained from the fit are
presented in Table 2. Because the population of Fe at
the 8d sites should be twice larger than that of Fe at each
of remaining sites, one concludes that the most intense
component S(1) corresponds to the Fe1 site. It was
observed in U-Fe-Al [7] and in Fe-Sn [6] alloys that
hyperfine magnetic field (HMF) of Fe increases with
a number of Fe NN. Assuming similar behavior in the
studied compound one concludes that S(2), S(3) and
Fig. 1. Mössbauer spectrum measured at room temperature.
Fig. 2. Mössbauer spectrum measured at T = 13 K.
Results and discussion
Hyperfine fields in UFe5Sn compound
S69
Table 2. The hyperfine parameters obtained from the spectrum measured at T = 13 K for different local surroundings (eQ
– electric quadrupole moment, Vzz – z component of electric field gradient, HMF – hyperfine magnetic field, I – intensities of
the components)
IS
(mm/s)
S(1)
S(4)
S(3)
S(2)
D
0.104 ± 0.005
0.094 ± 0.005
0.01 ± 0.03
0.23 ± 0.02
0.02 ± 0.02
eQVzz/2 (mm/s)
0.46 ± 0.02
−0.63 ± 0.05
0.49 ± 0.10
0.56 ± 0.05
0.59 ± 0.04
HMF
(T)
13.4 ± 0.2
14.6 ± 0.2
13.0 ± 0.3
7.9 ± 0.2
0
I
(%)
36 ± 2
22 ± 2
20 ± 4
19 ± 4
3±4
S(4) comes from Fe2, Fe3 and Fe4 sites, respectively,
which is consistent with [3].
In contrast to the results of [3], relative area of the
doublet D observed at low temperature is twice smaller
than that of the sextet (HMF = 2T) [3]. The origin of
this doublet is not clear because its relative area is small.
Comparison of hyperfine parameters obtained here
indicates that value of IS of the doublet D is close to
that of S(3) component, which in turn indicates that
this doublet may have Fe3 site origin, as suggested in
[3]. However, it is not clear, why part of Fe iron atoms,
located at the Fe3 site is still in the paramagnetic state
at T = 13 K. Another possibility is that the considered
doublet results from a small fraction of Fe atoms located
at antisite positions, U or Sn atoms.
Acknowledgment We are grateful to Prof. R. Troć for
the suggestion to undertake this study.
References
1. Akai H, Blugel S, Zeller R, Dedeichs PH (1986) Isomer
shifts and their relation to charge transfer in dilute Fe
alloys. Phys Rev Lett 56:2407−2410
2. Bergmann U, Shastri SD, Siddons DP, Batterman BW,
Hastings JB (1994) Temperature dependence of nuclear
forward scattering of synchrotron radiation in α-57Fe. Phys
Rev B 50:5957−5961
3. Gonçalves AP, Godinho V, Noel H (2000) Synthesis,
crystal structure, and magnetic properties of UFe5Sn. J
Solid State Chem 154:551−556
4. Gonçalves AP, Waerenborgh JC, Almeida M et al. (2003)
Magnetic properties of UFe5Sn single crystal. J Magn
Magn Mater 260:473−479
5. Margulies S, Ehrman JR (1961) Transmission and line
broadening of resonance radiation incident on a resonance absorber. Nucl Instrum Methods 12:131−137
6. Vincze I, Aldred AT (1974) Mössbauer measurements
in iron-base alloys with nontransition elements. Phys Rev
B 9:3845−3853
7. Waerenborgh JC, Gonçalves AP, Almeida M (1999)
Magnetization of the Fe sublattices in UFexAl12−x (4 ≤ x
≤ 5.8) studied by Mössbauer spectroscopy. Solid State
Commun 110:369−374